Google Data Center Efficiency Best Practices Full Video

Summary

Google's data centers, crucial for delivering web services, are designed for energy and resource efficiency. Innovations in design have reduced energy usage by half compared to typical centers. Key practices include optimizing Power Usage Effectiveness (PUE), managing airflow to prevent hot and cold air mixing, increasing cold aisle temperatures, utilizing free cooling, and minimizing power conversion stages. These strategies not only save millions in energy costs but are also applicable to data centers of all sizes, promoting sustainability.

Takeaways

• 🌐 Google prioritizes data center efficiency to deliver web services with minimal resource use.
• 🔍 Data centers can range from small closets to large warehouse-scale buildings filled with servers.
• 💡 Google has innovated in data center design to reduce energy, water, and other resource consumption.
• 🌡️ The ICT sector, including data centers, contributes about 2% to global greenhouse gas emissions, with data centers accounting for 0.15%.
• 🛠️ Energy-efficient design choices and best practices can significantly improve data center performance and reduce energy use.
• 📈 Google's efforts in data center efficiency have resulted in substantial energy savings, amounting to millions of dollars.
• 🔬 Measuring Power Usage Effectiveness (PUE) is crucial for managing data center efficiency and should be done continuously.
• 🚫 Proper air flow management, including eliminating hot and cold air mixing, is essential for reducing energy load.
• 🌡️ Increasing the temperature in the cold aisle can save energy costs and is supported by ASHRAE guidelines up to 80 degrees Fahrenheit.
• 🌡️ Free cooling, which utilizes ambient temperatures, can significantly reduce the need for energy-intensive cooling equipment.
• 💧 Evaporative cooling and sea water cooling are examples of efficient water use in data centers, leading to substantial water savings.
• 🔌 Minimizing power conversion stages and using efficient power supplies can save energy and reduce costs in data centers.
• 🔄 Implementing these efficiency practices is cost-effective and can be applied to both small and large data centers.

Q & A

Why are data centers crucial for Google's operations?

Data centers are essential for Google as they deliver all of their web services to users. They can range from small closets with a few machines to large-scale buildings optimized for power use and IT computing, housing thousands of servers.

What has Google done to innovate data center design and construction?

Google has spent a significant amount of time innovating the design and construction of their data centers to minimize energy, water, and other resource usage, focusing on sustainability and efficiency.

How much energy does the ICT sector contribute to global greenhouse gas emissions?

The ICT sector, which includes mobile phones, computers, monitors, cell phone towers, etc., contributes roughly about 2% of global greenhouse gas emissions.

What is the significance of the data center's energy usage in the ICT sector's total emissions?

The data center portion of the ICT sector is responsible for about 15% of the total emissions within the sector.

What is the Power Usage Effectiveness (PUE) and why is it important?

PUE is the ratio of total facility energy to IT equipment energy within a data center. It measures how effectively power and cooling are delivered to the IT equipment. A lower PUE indicates more efficient energy use and is desirable.

What was the typical PUE of an enterprise data center in 2006 and what has changed since then at Google?

In 2006, the typical PUE of an enterprise data center was 2.0. At Google, they have continuously improved their PUE, reaching a Time-Weighted Average (TTM) PUE of 1.16 over the last 12 months.

Why is it important to measure PUE frequently and over time?

Frequent and time-based measurement of PUE provides meaningful results that reflect the actual operation of the data center. Snapshots in time can be misleading and do not accurately represent the data center's efficiency.

How does Google manage PUE measurements in their data centers?

Google incorporates PUE measurements into their building management system across all sites, ensuring easy access to data and enabling efficient operation of their data centers.

What is the significance of managing air flow in a data center?

Managing air flow is crucial for reducing the data center's energy load. It involves eliminating the mixing of hot and cold air, which can be achieved through various containment approaches and CFD analysis.

How does Google use thermal modeling to improve air flow in their data centers?

Google uses thermal modeling to identify hot spots and understand how air flow is directed in their data centers. This helps them make simple design choices to improve air flow, such as using sheet metal extensions to increase intake to CRACs (Computer Room Air Conditioners).

What is the role of blanking panels in a data center and how do they contribute to energy savings?

Blanking panels are used in the rack space where not all equipment is populated. They help create a tight environment, similar to weatherizing a house, preventing air from bypassing the IT equipment and thus improving cooling efficiency and saving energy costs.

Why is raising the temperature in the cold aisle of a data center beneficial?

Raising the temperature in the cold aisle allows for more efficient cooling and reduces the need for heavy energy-consuming equipment like chillers. ASHRAE recommends running up to 80 degrees Fahrenheit, which Google implements, saving thousands of dollars in energy costs annually.

What is free cooling and how does Google utilize it in their data centers?

Free cooling is the practice of using ambient temperatures outside the data center for cooling without operating energy-intensive equipment. Google uses free cooling in all their data centers, which is reflected in their PUE data and results in significant efficiency gains.

How does Google implement evaporative cooling in their European data centers?

In Belgium, Google uses evaporative towers without chillers, taking advantage of local ambient conditions. In Finland, they use sea water cooling from the Bay of Finland, which cools the servers and then is tempered before returning to the bay, maintaining environmental balance.

What are some of the energy-efficient practices Google has implemented in their data centers?

Google has implemented several practices, including eliminating conversion stages in power supply, using battery on board trays, and ensuring AC-DC power supplies are energy efficient, which collectively save over 30% per year per server.

How do the efficiency best practices mentioned in the script apply to data centers of different sizes?

The efficiency best practices are applicable to both small and large data centers. They involve following five key steps to reduce energy use, making them a universal approach for improving data center efficiency.

Outlines

00:00

🌐 Energy Efficiency in Google's Data Centers

This paragraph discusses the significance of data centers at Google and their role in delivering web services. It highlights Google's efforts to innovate in data center design to reduce energy, water, and other resource usage. The speaker emphasizes the importance of adhering to best practices for energy efficiency, which have resulted in Google using half the energy of a typical data center. The paragraph introduces the concept of Power Usage Effectiveness (PUE) as a measure of energy efficiency and mentions Google's impressive PUE of 1.16, which indicates a highly efficient data center operation. The speaker also stresses the importance of continuous measurement for meaningful results and the incorporation of PUE measurements into building management systems.

05:01

🛠️ Best Practices for Data Center Efficiency

The second paragraph delves into specific best practices for improving data center efficiency. It starts with managing air flow to prevent the mixing of hot and cold air, using computational fluid dynamics (CFD) analysis to model and improve airflow. The use of cost-effective solutions like meat locker curtains and sheet metal doors to create hot and cold aisle containment is mentioned. The paragraph also addresses the misconception that data centers need to be kept cold, citing ASHRAE guidelines that recommend operating temperatures up to 80 degrees Fahrenheit, which Google follows. The benefits of free cooling, which utilizes ambient temperatures for cooling without heavy energy-consuming equipment, are discussed, along with the use of evaporative cooling and sea water cooling in Google's European data centers. The paragraph concludes with the importance of minimizing power conversion stages and using efficient power supplies to save energy, and it underscores the cost-effectiveness and applicability of these best practices to data centers of all sizes.

Keywords

💡Data Centers

Data Centers are facilities that house a large number of servers and networking equipment, crucial for delivering web services to users. In the video's context, they are central to Google's operations and are designed to be energy and resource-efficient. The script mentions that Google has innovated in the design and construction of these facilities to minimize their environmental impact.

💡Energy Efficiency

Energy Efficiency refers to the measure of how well a device, system, or process uses energy to deliver services. The video emphasizes Google's efforts to reduce energy consumption in their data centers, which is a key aspect of their sustainability practices. The script provides examples of how Google has achieved this, such as optimizing power usage effectiveness (PUE).

💡Power Usage Effectiveness (PUE)

PUE is a metric that indicates the efficiency of a data center's power usage, calculated as the ratio of total facility energy to IT equipment energy. A lower PUE indicates more efficient power and cooling delivery to IT equipment. The script explains that Google aims to reduce PUE to as close to 1.0 as possible, which represents ideal efficiency.

💡ICT Sector

ICT stands for Information and Communication Technology. The script mentions that the ICT sector, which includes devices like mobile phones and infrastructure like cell phone towers, contributes about 2% to global greenhouse gas emissions. Data centers are a significant part of this sector, and their energy efficiency is critical to reducing the sector's environmental impact.

💡Airflow Management

Airflow Management is the process of controlling the movement of air in a data center to ensure optimal cooling of IT equipment. The script discusses the importance of eliminating the mixing of hot and cold air for efficient cooling. Google uses techniques like CFD analysis and physical barriers to manage airflow effectively.

💡Hot and Cold Aisles

In data center design, hot aisles and cold aisles refer to the organization of racks to separate hot air exhaust from cold air intake, enhancing cooling efficiency. The script describes how Google uses curtains and sheet metal doors to create physical separation between hot and cold aisles, improving airflow and reducing energy use.

💡Blanking Panels

Blanking Panels are used in data centers to cover empty spaces in server racks, preventing the mixing of hot and cold air and improving cooling efficiency. The script mentions that Google uses these panels to 'weatherize' their data centers, likening the practice to sealing gaps in a house to improve energy efficiency.

💡Free Cooling

Free Cooling is a method of using outside ambient temperatures to cool a data center without the need for energy-intensive mechanical cooling systems like chillers. The script explains that Google utilizes free cooling at all of its data centers, which results in significant energy savings and efficiency gains.

💡Evaporative Cooling

Evaporative Cooling is a technique that uses the evaporation of water to cool air, which is then used to cool IT equipment in a data center. The script provides examples of Google's use of evaporative cooling in Belgium and Finland, highlighting the water and energy savings achieved through this method.

💡UPS System

UPS stands for Uninterruptible Power Supply, a system that provides emergency power to a data center in the event of a power outage. The script discusses the energy losses associated with traditional UPS systems due to multiple conversion stages between AC and DC power. Google has innovated by implementing battery on board trays to eliminate these conversion stages, increasing efficiency.

💡AC-DC Power Supply

AC-DC Power Supply refers to the conversion of alternating current (AC) to direct current (DC), which is required by most IT equipment. The script mentions that by ensuring the efficiency of AC-DC power supplies, significant energy savings can be achieved. Google saves over 30 percent per year per server by implementing efficient power supplies.

Highlights

Google data centers are critical for delivering web services.

Data centers range from small closets with a few machines to large warehouse-scale buildings.

Google has innovated the design and construction of data centers to minimize energy, water, and resource use.

Google's data centers use half the energy of typical data centers.

The ICT sector is responsible for about 2% of global greenhouse gas emissions, with data centers contributing 15% of that.

Implementing best practices in energy efficiency can significantly improve data center performance.

Google's energy efficiency practices save millions of dollars annually.

Power Usage Effectiveness (PUE) is crucial for measuring data center efficiency, with Google achieving a PUE of 1.16.

Regular PUE measurements are important for accurate and meaningful results.

Airflow management, particularly eliminating the mixing of hot and cold air, is key to reducing energy load.

CFD analysis helps model airflow and make design improvements in data centers.

Simple retrofits, like meat locker curtains and sheet metal doors, can significantly enhance airflow management.

Raising cold aisle temperatures to recommended levels (up to 80°F) can save thousands of dollars in energy costs.

Free cooling, utilizing ambient temperatures, can yield substantial efficiency gains.

Eliminating unnecessary power conversion stages and using efficient components can reduce energy consumption.

Google's on-board battery approach eliminates multiple conversion steps, enhancing energy efficiency.

Efficient power supplies, such as those with Energy Star labels, can further reduce energy use.

Google saves over $30 per server annually through these efficiency measures.

Many efficiency practices are cost-effective within 12 months and applicable to both small and large data centers.

Transcripts

[Music]

here at google

data centers are very very important to

us they are how we deliver all of our

web services to

all of our users and data center can

mean a variety of things it can mean a

small closet filled with a couple of

machines all the way to very large

warehouse scale

buildings that are optimized for power

use and

i.t computing and filled with thousands

of servers and google we've spent a lot

of time

innovating the way in which we design

and build these facilities

to minimize the amount of energy and

water and other resources that these

computing facilities use

in terms of the results of all of the

work that we've been doing

over many many years now who use half

the energy

of the typical data center to put things

into perspective the entire ict sector

that includes mobile phones computers

monitors cell phone towers

is roughly about two percent of global

greenhouse gas emissions

of that two percent the data center

portion is responsible for about 15

there's design choices that you can make

for energy efficiency that improve

right the performance of your data

center and these things are just best

practices and adhering

well to best practices that's how you

can actually make the most improvements

in terms of energy use

the results of these types of activities

return google

millions of dollars in energy savings so

the results are significant

we've invited several members of our

data center team here to explain some of

these best practices

to all of you

the first step in managing the

efficiency of your data center is to

make sure you have the instrumentation

in place

to measure the pue or power usage

effectiveness

pue is the ratio of total facility

energy to i.t equipment energy within

your data center

it's a measure of how effectively you

deliver power and cooling to the it

equipment

in 2006 the typical pue of an enterprise

data center was 2.0 which means

for every one watt of i.t energy

consumed one watt of overhead was

consumed by the facility to deliver the

power and cooling

reducing the overhead is really what you

want you want pue to get to as close to

1.0 as possible

over the last 12 months our ttm pue was

1.16

we've continuously measured that and has

gone down nearly every quarter since we

began reporting it back in 2008.

last quarter the lowest data center was

1.09

ideally you should measure it as fast as

you can as often as you can

every second or so and the more often

you can measure it the

more meaningful the results will be it's

important to

you measure pue over the course of a

year annually or quarterly

to get meaningful results if you just

take snapshots in time

the information won't be realistic and

it won't

really be an actual measure of how well

your data center is operating

one way to make it easier to manage is

incorporate the pue measurements into

your building management system

we do this all of our sites at google

without having easy access to this data

we wouldn't be able to operate our data

centers as efficiently as we do

once you have the ability to measure and

manage your pue the first step in terms

of reducing your data center energy load

is to focus on the management of the air

flow

the most important thing here is to

eliminate the mixing of the hot and the

cold

air and there's no one right way to do

this containment can be achieved through

many different approaches

one thing that we found very useful at

google is cfd analysis to see

where are your hot spots and how is your

airflow going to actually

be directed in your data center by doing

so you can actually model the way in

which airflow will go and it helps you

make very simple design choices to

improve the airflow in your data center

for example in one of our computing and

networking rooms we call them cnrs

we actually did some thermal modeling to

see exactly what airflow was doing

through that modeling we realized that

the intake to our cracks was too low

and that by simply piecing together some

sheet metal we could create extensions

it would dramatically increase the

airflow quality

into the cracks we also did a bunch of

other retrofits

here in this corporate data center at

google we've implemented

meat locker curtains that are very

inexpensive and easy to install

these are hung from the overhead

structure and they separate

the cold aisle which is actually hot in

the hot aisle

which is actually hotter we are set now

to enter the hot aisle containment

door and we we incorporated the simple

inexpensive sheet metal doors

to separate very tightly

the cold aisle from the hot aisle now

over here

we've got the hot air from the racks are

coming up

going overhead up through the returner

plenum back to the crack units to give

you a nice high

temperature differential across your

pregnancy

a very important step is to

the rack space where you don't quite

have all of your equipment populated and

it's very easy to do

with these blanking panels it's almost

like weatherizing your house

make sure that you've got a nice tight

environment

all told we spent about twenty five

thousand dollars in parts and those

twenty five thousand dollars saved us

over sixty five thousand dollars in

energy costs

yearly

[Music]

once you manage your airflow properly

the next step in data center efficiency

is to increase the temperature of your

cold aisle

it's long been believed by many data

center operators that the data center

has to be cold

to keep all the equipment at a

temperature that it will run safely at

and in fact that's just false so if you

look at recommended guidelines from

ashrae

they recommend you running all the way

up to 80 degrees fahrenheit

and at google that's exactly what we do

we've got a small

corporate data center here it's about

200 kilowatts of load

simply raising the temperature from 72

degrees to 80 degrees

saves us thousands of dollars in energy

costs every single year

what's nice about that is it also allows

our employees to come to work

in shorts

[Music]

whenever possible we'd recommend people

to free cool

pre-cool means utilizing ambient

temperatures outside of your data center

to be able to provide cooling without

operating very heavy energy consuming

equipment like chillers

we use free cooling at all of our data

centers and you can see this in our

publicly reported pue data where the pub

values go up

in the summer time and down in the

winter time and this is just a

reality of running your our operations

with

free cooling and it yields tremendous

efficiency gains

in europe we have two data centers that

have no chillers

whatsoever we're able to take advantage

of the local constraints and

conditions in belgium we use evaporative

towers

without any chillers given the ambient

conditions

in finland we use sea water cooling sea

water from the bay of finland

cools the servers and then we temper the

water returning to the bay of finland so

there's no

temperature gradients returning to the

bay evaporative cooling

uses water on site but what we found

through our studies is that

by the use of evaporative cooling in a

very efficient fashion we save water

on the whole so for every gallon of

water we use in the evaporative cooling

plants

we eliminate the use of two gallons of

water on the energy production side

so this translates into hundreds of

millions of gallons per year in water

savings

there's no one right way to deliver free

cooling

the important point is that you should

examine these opportunities and take

advantage of them

to eliminate or reduce substantially the

mechanical cooling

in a data center you pull power in from

the electrical grid

and you convert it down to the voltages

that are needed for all the components

in the data center

and there's a lot of conversion stages

in there by minimizing those conversion

stages you can

save money and save energy also by

making each conversion stage more

efficient

you can save energy as well

traditionally one of the

the big losses is the ups system the

uninterruptible power supply

typically there's a giant room of

batteries batteries are dc voltage and

the power coming in to charge those

batteries is ac

and so you need to convert the ac down

to dc with a rectifier

in order to charge the batteries and

then when the batteries are needed in a

power event you need to convert that

back to ac with an inverter

and then the ac needs to be converted

back down to dc for all the components

in the data center

so you've got three conversion stages in

there that are not necessary

what google has done is put a battery on

board the tray so you're eliminating

those three conversion steps you just

have dc right into the server components

in a typical server configuration you

have a server with an ac dc power supply

attached to it by making sure that ac dc

power supply is efficient

you can save a lot of energy things like

energy star labels will point you to

power supplies that are 90 plus percent

efficient

google is able to save over 30 per year

per server by implementing all of these

features

there really are very simple effective

approaches that all of us can implement

to reduce

the data center energy use and most of

them are cost effective within 12 months

of operation

so a lot of efficiency best practices

should be adopted by just about everyone

they're applicable to small data centers

or large data centers

it's simply following the five steps

that we go through here

to make sure that you're able to reduce

your energy use

[Music]

you